The deregulation of the electric power industry has transformed electricity into a commodity the price of which follows supply and demand imbalances. We consider a stochastic extension to the problem of scheduling the generation units of a power company, also known as the unit
commitment problem. The suggested model captures uncertainty in electric load, electricity price, fuel price, and fuel constraints. Uncertainty is modeled using a set of scenarios that reflect varying market demand and
prices. We solve the resulting stochastic program using Lagrangian relaxation. Extensions, such as auctioning excess short-term capacity, will be discussed.

Speaker Bio:

Samer Takriti manages the Stochastic Optimization group within the Mathematical Sciences Department at IBM TJ Watson Research Center in Yorktown Heights, NY. He holds a PhD degree in Industrial and Operations Engineering from the University of Michigan. Prior to joining IBM, he was a Director of Research at Enron Corporation where he headed the
quantitative support for Enron Broadband Services. His research focuses on stochastic programming with applications to the modeling of energy and natural resources. He is the president of INFORMS Section on Energy, Natural Resources, and the Environment.